1. Home
  2. Field-Programmable Gate Arrays
  3. Xilinx 6.2i ISE WebPACK running under wine?
  4. Page 2

Xilinx 6.2i ISE WebPACK running under wine?

The "conventional wisdom" for some time has been that the command line tools, which are the things that do the actual work, were written on Unix platforms. After all, I seem to remember a time (my memory could be faulty here) when the tools were only available on Unix. In those days (not really so long ago), the GUIs were also written in X (probably Motif, which is what most vendors were using at the time).

It looks to me like the command line tools are still either developed on Unix, or on Windows/Unix pretty much simultaneously. Only the gui has become a Windows thingy. I mainly think this because it is clear that the command line tool interface really has not changed much since "the old days", and they remain separate from the gui even now. The gui remains just a button clicking front end to those tools. I have not used Solaris in awhile, so I have no idea how that GUI performs these days.

--
My real email is akamail.com@dclark (or something like that).
Reply to
Duane Clark
Loading thread data ...

Andrew Rogers wrote: : Uwe Bonnes wrote: : > Phil Tomson wrote: : > ,,, : > : That still won't help when it comes to programming devices. The Jungo : > : parallel port driver doesn't work under Wine. : > : > I had some success with

formatting link
running with wine. : > Nahitafu will hopefully enable programming of more devices.. : > : > Bye

: How about GtkJTAG from the same author? That runs on Linux, I have : downloaded and compiled it. The downside is that in needs to run as root

As I understand Nahitafu, the Gtk Version is abandoned. Chip manufactures only prelease the needed programming information, if Nahitafu doesn't open up the code.

The windows version running on wine doesn't need root access, if the ppdev/parport device is accessible to the user and ppdev configured right in wine.

Bye

--
Uwe Bonnes                bon@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------
Reply to
Uwe Bonnes

I now have GtkJTAG working without being root. I am using the parport driver. I think you're right about GtkJTAG being abandoned by Nahitafu, good job I have the source code under GPL license:).

I will release my version as soon as I can verify it works when my Spartan3 kit arrives.

GtkJTAG understands Xilinx BSDL files so there shouldn't be a problem.

The secrets seem to be in the bitstream encoding, how bitgen works? I had started to work on my own bitgen a while ago. I made some progress by taking a very simple design. If I remember correctly I did an auto-correlation on the bitstream. One of the interesting things I found was that the bitstream flipped half way through. If there is anyone interested in writing their own bitgen contact me and I will find my previous work.

Regards Andrew

Reply to
Andrew Rogers

: I now have GtkJTAG working without being root. I am using the parport : driver. I think you're right about GtkJTAG being abandoned by Nahitafu, : good job I have the source code under GPL license:).

: I will release my version as soon as I can verify it works when my : Spartan3 kit arrives.

Sound's interesting.

: GtkJTAG understands Xilinx BSDL files so there shouldn't be a problem.

: The secrets seem to be in the bitstream encoding, how bitgen works? I : had started to work on my own bitgen a while ago. I made some progress : by taking a very simple design. If I remember correctly I did an : auto-correlation on the bitstream. One of the interesting things I found : was that the bitstream flipped half way through. If there is anyone : interested in writing their own bitgen contact me and I will find my : previous work.

Sounds even more interesing...

--
Uwe Bonnes                bon@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------
Reply to
Uwe Bonnes

I believe that your right. The real tools like map and par, and now xst, were always developed on Unix and ported to Windows. Those tools are straight POSIX C or C++ so they will run on anything with just a recompile. They always ran under WINE without a hitch and the Linux native versions run on any flavor of Linux. The GUI tools are another story. They are written for Windows and ported to Linux using a terrible tool that has all sorts of library dependencies and awful performance. Both the Cadence and Mentor tools will run on any flavor of Linux, I'm using them with both Mandrake 9.2 and 10.0, so it clearly possible to write a GUI that doesn't have any dependentcies on a particular distribution. Also I run Cadance and Mentor tools over a network without any performance issues, both Simvision and ModelSim are true X applications that work flawlessly when split between a client on one machine and an X server on another.

The Xilinx GUI tools have very real distribution dependentcies which shows that the people responsible for porting them don't have a good understanding of the Linux world. Also they are unusable over a network which is their Windows heritage rearing it's ugly head. The GUIs will not run on Mandrake 10.0 although they do run fine on Mandrake 9.2. Impact, which requires a kernel patch for some reason, only works on Redhat 8.0. They provide the kernel patch but it's specific to the kernel used in RH8, I think it's 2.4.17, which is hopelessly obsolete. They haven't bothered to provide the patch for modern kernels.

I've upgraded my servers to Mandrake 10.0 (2.6 kernel) because I do everything from scripts so I don't care if the GUI tools work. I'm keeping my workstation at Mandrake 9.2 so that I can use FPGA Editor if I need to.

Reply to
General Schvantzkoph

I recompiled the kernel patch on RH9 (kernal 2.4.20). Impact is working fine for me.

--
My real email is akamail.com@dclark (or something like that).
Reply to
Duane Clark

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.